A Record of Environmental Change in Caribbean Coral Reefs: Sclerochronology and Geochemistry of O. faveolata as a Paleoclimate Proxy at Coral Gardens and Rocky Point, Belize

Abstract:

Over the last several decades, Acropora cervicornis has seen a massive die-off in the Caribbean. (Aronson and Precht 2001; Gardner et al., 2003; Greer et al., 2009). The potential causes of decline in A. cervicornis in the Caribbean include: extremes in sea surface temperatures (SST), ocean acidification, eutrophication, white-band disease, storm disturbances, and other anthropogenic disturbances. Contrary to the regional decline in A. cervicornis, Coral Gardens on the Belize Barrier Reef has an Acropora sp. population that appears to be thriving. Through a combination of sclerochronology, stable isotope analysis, and in situ sensor data, this work capitilizes on the opportunity to study reef conditions in a site where micro-environmental conditions appear to be favorable for healthy A. cervicornis coral growth. We use cores from two Orbicella faveolata colonies located within Acropora stands, as A. cervicornis does not reveal annual banding. We compare two cores from one O. faveolata colony at Coral Gardens, first cored in 2011 and again in 2014, to one O. faveolata core at near-by Rocky Point, where A. cervicornis is much less abundant. These cores were x-radiographed in order to expose the annual banding and sampled for stable oxygen and carbon isotope analysis (10-15 samples/cm). We show that, although there are no significant differences in the range of the δ¹⁸O and δ¹³C signature between Rocky Point and Coral Gardens, there is a clear difference in the stress histories at these locations as inferred from linear extension rates (LER’s) and annual banding patterns. Rocky Point averages a LER of 10.5±1.4 mm/year (n = 29) over a 30 year record and Coral Gardens averages 9.1±1.2 mm/year (n = 70) from ~1953 – 2001, and averages 6.2±1.6 mm/year (n = 36) from coral years 2001-2014 after an inferred stress-banding event. This is in contrast to the observed overall health of A. cervicornis at the two locations. The inferred stress-banding event is currently unidentified but is most likely due to either SST stresses during the 1998 El Niño summer or damage as a result of Hurricane Keith in 2000. Finally, we show that there may be a long-term trend in declining LER’s (-4.4 mm/year, R² = 0.29) at Coral Gardens prior to the inferred stress-banding event consistent with work on the Great Barrier Reef (De’ath et al., 2009).